From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S261208AbVFBS2e (ORCPT ); Thu, 2 Jun 2005 14:28:34 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S261218AbVFBS2d (ORCPT ); Thu, 2 Jun 2005 14:28:33 -0400 Received: from one.firstfloor.org ([213.235.205.2]:3991 "EHLO one.firstfloor.org") by vger.kernel.org with ESMTP id S261208AbVFBS2a (ORCPT ); Thu, 2 Jun 2005 14:28:30 -0400 To: "Martin J. Bligh" Cc: jschopp@austin.ibm.com, Mel Gorman , linux-mm@kvack.org, linux-kernel@vger.kernel.org, akpm@osdl.org Subject: Re: Avoiding external fragmentation with a placement policy Version 12 References: <20050531112048.D2511E57A@skynet.csn.ul.ie> <429E20B6.2000907@austin.ibm.com> <429E4023.2010308@yahoo.com.au> <423970000.1117668514@flay> <429E483D.8010106@yahoo.com.au> <434510000.1117670555@flay> From: Andi Kleen Date: Thu, 02 Jun 2005 20:28:26 +0200 In-Reply-To: <434510000.1117670555@flay> (Martin J. Bligh's message of "Wed, 01 Jun 2005 17:02:35 -0700") Message-ID: User-Agent: Gnus/5.110002 (No Gnus v0.2) Emacs/21.3 (gnu/linux) MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org "Martin J. Bligh" writes: > It gets very messy when CIFS requires a large buffer to write back > to disk in order to free memory ... How about just fixing CIFS to submit memory page by page? The network stack below it supports that just fine and the VFS above it does anyways, so it doesnt make much sense that CIFS sitting below them uses larger buffers. > There's one example ... we can probably work around it if we try hard > enough. However, the fundamental question becomes "do we support higher > order allocs, or not?". If not fine ... but we ought to quit pretending > we do. If so, then we need to make them more reliable. My understanding was that the deal was that order 1 is supposed to work but somewhat slower, and bigger orders are supposed to work at boot up time. -Andi